MAGNETOM Amira; MAGNETOM Sempra

The MAGNETOM Amira and MAGNETOM Sempra systems, enhanced with syngo MR XA50M software, are magnetic resonance diagnostic devices that leverage new dedicated coils for the Sempra model and a suite of advanced software, including AI-powered image reconstruction like Deep Resolve. These systems aim to improve image quality through techniques trained on extensive datasets, expand diagnostic capabilities with novel imaging sequences, and optimize clinical workflow for comprehensive anatomical and functional assessment across various body regions.

The MAGNETOM Amira and MAGNETOM Sempra systems, with the updated syngo MR XA50M software, are magnetic resonance diagnostic devices (MRDD) designed to produce a variety of images, including transverse, sagittal, coronal, and oblique cross-sections, along with spectroscopic images and spectra. Their purpose is to display the internal structure and function of the head, body, or extremities, with contrast agents usable as needed for specific regions of interest. The resulting images, spectra, and derived physical parameters are intended to assist trained physicians in diagnosis and can also support imaging during MR-compatible interventional procedures. The update introduces new hardware specific to the MAGNETOM Sempra: the Flex Large 8 Coil, Flex Small 8 Coil, and Flex 8 Coil Interface. Both systems incorporate numerous new and modified software features, prominently featuring AI-driven image enhancement tools such as Deep Resolve Sharp and Deep Resolve Boost. These AI features were developed using extensive datasets (e.g., Deep Resolve Boost: 26,473 2D slices; Deep Resolve Sharp: 13,977 2D slices) from in-house and collaborative measurements on 1.5T and 3T scanners, covering various body regions and protocols, and validated using metrics like PSNR and SSIM, with visual inspection and clinical evaluations in peer-reviewed studies showing beneficial use. Other software enhancements include advanced sequences (SMS TSE DIXON, SE EPI MRE, WAVE-CAIPI SWI), workflow improvements (myExam Brain Autopilot, Access-i), and motion correction (TSE MoCo). Non-clinical tests, including sample clinical images and software verification, demonstrated the systems perform as intended.

AI Regulatory Concerns

Powered by

Jev reads the full FDA letter and returns calibrated concern probabilities per regulatory category. These are AI estimates, not regulatory advice.

• Q1. How do the 'Deep Resolve Sharp' and 'Deep Resolve Boost' AI algorithms specifically impact diagnostic accuracy or reader confidence across different anatomical regions and patient populations, beyond the reported PSNR and SSIM improvements?; What are the quantitative benefits (e.g., scan time reduction, improved signal-to-noise ratio) of the new Flex coils (Flex Large 8, Flex Small 8) for the MAGNETOM Sempra in common clinical scenarios?; Can the 'myExam Brain Autopilot' and 'Automatic fiducial detection' features adapt to atypical anatomies or significant pathological changes, and what is the failure rate or need for manual override in such cases?; How does the performance of 'SE EPI MRE' (Magnetic Resonance Elastography) on these systems compare to established MRE techniques on other platforms, particularly in terms of stiffness value accuracy and reproducibility?; What are the specific computational requirements and processing times for the new AI-driven reconstruction features (Deep Resolve suite), and how might this impact clinical workflow in high-throughput environments?